Contents
01Questions about the people
Agencies sell with senior people and deliver with whoever is free. The fix is simple: ask to meet the engineer who will do the work, before the contract. Ask whether any part is subcontracted and to whom. Ask what happens if that person leaves mid project.
None of these are hostile questions. A team that works this way answers them in a sentence.
02Questions about ownership
Ownership disputes are the most expensive surprise in this industry, and they are entirely avoidable at the start.
- Whose account holds the repository during the project, and whose holds it after.
- Who owns the ad accounts, the analytics property and the domain.
- What you receive at handover beyond the code: documentation, credentials, a runbook.
- Whether anything stops working if the relationship ends.
If the answer to any of these is we will sort that out later, sort it out now instead.
03Questions about money
Ask how estimates are made and what happens when one is wrong, because one will be. A clear rule beats a confident promise: we absorb the first 15% of an overrun, then stop and re-plan with the client. Ask what the second year costs, including hosting and maintenance. Ask what is explicitly not included, which is the single most useful question on this list.
04Answers that should end the conversation
A few answers are reliable warnings, in any market and at any price point.
- Guaranteed rankings, guaranteed lead volume or a guaranteed conversion rate. Nobody controls those, and the guarantee is priced into the invoice.
- A quote given before anyone has looked at your data or your existing system.
- Case studies with numbers but no stated source for them.
- Reluctance to put the scope in writing because it slows things down.
- No work they will refuse. It means the only filter is your budget.
05What to check yourself in ten minutes
Open their own site on a phone and watch how long it takes to load. Look for published prices rather than a contact form. Check whether their case studies name the system each number came from. Send an enquiry and time the reply.
A company that cannot run its own acquisition well is an odd choice to run yours.
Questions
Is a cheaper offshore team a bad idea?
Not inherently. The risk is not location, it is the distance between the person who sells and the person who builds. Ask the same questions about people and ownership and judge the answers, not the time zone.
What should a discovery document contain?
What the system has to do, what it will not do, the integrations involved, the risks found while reading your data, and a price band with the factors that move it. If you could not hand it to another vendor and get a comparable quote, it is not a discovery document.
How fast should a serious vendor reply?
Within a business hour or two. More important is what the reply says: a vendor who tells you in the first message that they cannot start for six weeks is worth more than one who discovers it after a discovery call.
Where we can help